ABSTRACT Heart failure prevalence is increasing in the U.S., particularly among blacks and older adults. Heart failure is a progressive disease defined by stages, and onset of asymptomatic, preclinical heart failure is often evident decades before symptomatic, clinical disease. This provides an opportunity to discover novel targets for intervention during preclinical stages to prevent or attenuate progression to clinical heart failure stages. This early prevention approach is critical since, once an individual advances to the next stage, regression is unlikely. Reported moderate to vigorous intensity physical activity (MVPA) and cardiorespiratory fitness (CRF) are independently related to a reduced risk of lifetime, clinical heart failure; a relation that has been primarily studied in midlife or older adults. Nothing is known about the relations of light intensity physical activity (LPA) or sedentary behaviors (SED) with heart failure. This focus on event-driven endpoints discounts the progressive nature of heart failure and biases the field against discovery of novel targets for improving more proximal outcomes that are predictive of future clinical disease, including N-terminal pro-brain natriuretic peptide (NT-proBNP), high sensitivity cardiac Troponin T (hscTnT), peak oxygen uptake (peak VO2), and heart failure stage classification. Evidence from CARDIA demonstrates an increased prevalence of preclinical heart failure, and progression to more severe heart failure stages over 25-years defining young adulthood to midlife. This has occurred in parallel with declines in reported MVPA and CRF over these 25-years, and replacement of active time for sedentary time (based on accelerometry) during the first 10-years of midlife. Given that these biomarkers are not currently available in CARDIA, the impacts of these exposures have not been tested in this cohort, or in other studies. To address this need, we propose the CARDIA Activity and Heart Failure (ACT-HF) Study, a four-year ancillary study to the Year 35 core exam (2020-21; cohort ages 53-65 years). Participants will be all those who attend the core exam, who meet eligibility criteria, and agree to participate (estimated n?2,431). To complement 35-years of extant data, CARDIA ACT-HF measures include: (1) third accelerometry measures to fully characterize midlife, (2) first measures of NT-proBNP and hscTnT in CARDIA (at Years 20, 30 & 35 via stored blood samples), and (3) a final maximal graded exercise test (GXT) test for measured or estimated peak VO2. The validity of the 400-meter walk test, as a possible replacement for the maximal GXT in future exams, will be tested. We aim to examine the: (1) independent and simultaneous longitudinal relations of 20-year changes in a) reported MVPA and b) CRF from early adulthood to midlife with 15-year changes in heart failure biomarkers collected across midlife; (2) independent and simultaneous longitudinal relations of accelerometer-based a) MVPA, b) LPA, and c) SED changes with heart failure biomarkers across midlife; and (3) bidirectional relations of a) accelerometer-based MVPA, LPA, and SED and b) CRF with heart failure stages across midlife. Interactions of these relations by race and sex will be tested in all study aims.